On Image Analysis by the Methods of Moments
IEEE Transactions on Pattern Analysis and Machine Intelligence
A Theory for Multiresolution Signal Decomposition: The Wavelet Representation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Orthogonal Moment Features for Use With Parametric and Non-Parametric Classifiers
IEEE Transactions on Pattern Analysis and Machine Intelligence
3D zernike descriptors for content based shape retrieval
SM '03 Proceedings of the eighth ACM symposium on Solid modeling and applications
Efficient contour-based shape representation and matching
MIR '03 Proceedings of the 5th ACM SIGMM international workshop on Multimedia information retrieval
Aircraft identification by moment invariants
IEEE Transactions on Computers
Journal of Computer and System Sciences
Combining approaches for early diagnosis of breast diseases using thermal imaging
International Journal of Innovative Computing and Applications
Aircraft recognition using modular extreme learning machine
Neurocomputing
Hi-index | 0.00 |
Automatic Target Recognition (ATR) of infrared object has been taking a great interest to the researchers in recent years. ATR requires invariance of high cognition accuracy in translation, scaling and orientation, but classification of two-dimensional (2D) shapes despite of their position, size and orientation in infrared image remains a difficult problem. In this paper, a feature extraction method is proposed using Wavelet Moment Invariants (WMI). The very similar objects can be classified correctly by virtue of the wavelet moment with its multi-resolution properties. Compared with some other geometry moments, the classification rate and the recognition efficiency are improved with wavelet moments. As different wavelet basis will have different impacts to wavelet moment, it affects the efficiency of classification. Some important properties such as orthonomality, supported length and vanishing moments which affect the performance of wavelet moment are discussed in this paper. Through experimental analysis, a conclusion is obtained that symmetry, compactly supported wavelet has more high-performance, and using wavelet function with proper vanishing moments could effectively improve the efficiency of classification.